Another obvious reference on context-awareness is the work of Anind Dey. Feel free to contact me directly for more specific references.  

https://scholar.google.be/citations?user=ydA8Q5AAAAAJ

I recommend you to take a look at the work of Manfred Tscheligi:

http://dl.acm.org/results.cfm?query=Manfred%20Tscheligi&filtered=persons%2Eauthors%2EsearchDspName=Tscheligi%2C%20Manfred&within=owners%2Eowner%3DHOSTED&dte=&bfr=&srt=_score

My previous work focused on adaptation, besides context-awareness and UI design. In my Ph.D. thesis, I discuss a large number of references. You may have access within UCL's digital library.  

I would suggest checking on Abowd's work on context (although most references are not recent, he has a large body of work on 'context' n=34), and Bill Schilit's work too.

For more recent publications, you could also check Denzil's work (focus on mobile apps and a development framework 'Aware'): http://www.denzilferreira.com/publications/

Please let know if you need further information, or if you would like to discuss these research topics. 

http://www.gregoryabowd.com/abowd-search/WHhuUTY5MjUxX3MtRElTVElOQ1QobiZuaWQpX2ppbm5lci1uciZuaWQjbi5uaWRfamxlZnQtdG4mbmlkI24ubmlkX3ctbiZzdGF0dXM9MV93LShMT1dFUihuciZib2R5KSNsaWtlIyclJWNvbnRleHQlJScjT1IjTE9XRVIobnImdGl0bGUpI2xpa2UjJyUlY29udGV4dCUlJylfdy1uJnR5cGU9J3B1YmxpY2F0aW9uJw%3D%3D

The research domain of context-aware user interfaces can be addressed from multiple viewpoints. A wide body of literature discusses context-aware mobile systems. In such systems, location-based adaptations are commonly considered. If you are interested in this, you could also have a look at Seng Loke (2007) Context-Aware Pervasive Systems: Architectures for a new breed of applications. As the title indicates, this book mainly discusses the technical aspects of context-aware systems.

Personally, I am more interested in the HMI aspects of context-aware adaptable or self-adaptive systems. In my research, I study the effect of context-aware adaptable user interfaces on operators’ situation awareness in traffic management control rooms. Models developed for mobile systems are less suitable in this context. In terms of definitions of context-aware systems I would indeed refer to Dey. I, however, did not yet come across a classical work in terms of addressing HMI aspects of context-aware user interfaces. Instead, I look at literature addressing HMI aspects of automation, especially when addressing adaptive interfaces. For example “Parasuraman, R., Sheridan, T.B., & Wickens, C.D. (2000) A model of types and levels of human interaction with automation, IEEE Transactions on Systems, Man, and Cybernetics – Part A, 30 (3), 286-297”

Context awareness is not my focal area of interest. Thus I would not venture to name the seminal papers in this area.  However, I am compelled to mention that the concept of building context aware interfaces and efforts to address the design issues  may go as far back as the late 70s / early 80's (perhaps further).  During that period, US military human factors laboratories such as AFAMRL were exploring HMI methods to reduce pilot mental (cognitive) workload created by the complexities of new missions, weapons system capabilities and most importantly the veritable flood of new information that it was possible to deliver to the cockpit.  One of the key solutions first studied at that time was to optimize operator loading through a combination of adaptive displays and adaptive automation support.   Thus, the concept  was to provide displays could be reorganized, and receive filtered information and to introduce partial automation support (e.g. 1 axis of navigational control) based on contextual factors consisting of  current mission phase, threats detected, and the psychophysiological status of the pilot.   

To put it briefly, our recent technology developments have made context aware systems a more tangible reality.  However, current researchers and designers may find themselves reinventing the wheel" if they ignore early research conducted on the (non-technology dependent) problems of building context aware displays and automation supports that are effective in enhancing or maintaining human-machine system performance